Georges Haour & Laurent Mieville
from SCIENCE to BUSINESS How Firms Create Value by partnering with Universities
2
CONTENTS
List of Figures Preface Chapter 1 Firms engage with universities in many different ways Universities and firms, two key actors of the so-called ‘knowledge economy’ Key is not how much firms invest in R&D but how they perform it Ways in which firms and universities engage with each other The graduates are key agents for transferring knowledge and technology University–industry consortiums for graduate/ continuing education Broad linkage between business and universities Professional contacts Focus of this book: collaborative research, licensing and spinning out start-ups The two themes running through this book Chapter 2 Collaborative research between companies and universities Unilateral firm–university collaboration Students in the firm’s R&D Multilateral collaborative projects Long-term consortia Co-location The innovation campuses of companies Joint laboratories
xi xiii 1 2 4 6 7 10 10 11 12 14 21 22 29 30 34 36 37 39
vii
9780230236516_01_prexiv.indd vii
9/9/2010 2:58:34 PM
Contents
viii
Embedded laboratories Endowments Secondment The weight of collaborative research in universities Cornell University’s policy on firms engaging with the university Collaborations in non-technical areas Conclusion Chapter 3 Firms accessing university technology through licenses IP-based licensing Licensing at Stanford University Elements of an effective technology licensing office The business of IP: patent aggregators, patent trolls, and patent pirates Losing sight of common sense in granting patents for licensing The difficulty of licensing public research: the case of BTG Universities need clear ownership of intellectual property Developing Technology Licensing Offices in Japan Licensing in the German system for technology transfer How about China? Turkey Conclusion Chapter 4 Firms accessing university research results via spin-outs Example of a university spin-out The incubation process The role of incubators The extravagant bubble of the dotcom start-ups
9780230236516_01_prexiv.indd viii
40 42 44 45 46 47 49 51 52 54 58 62 66 67 68 70 72 74 78 79 81 81 82 86 88
9/9/2010 2:58:35 PM
Contents From science to business at Imperial Innovations Incubating non-technical ventures How effective is the spinning out from universities? The overall scene in the UK The Peter Pan complex of Europe’s young companies Spin-outs at the University of Tokyo Science parks in China From route 128 to Silicon Valley and Bangalore’s Silicon plateau Entrepreneurial Israel Spin out or not spin out? this is the question Chapter 5 SMEs must engage with universities SMEs and competitiveness SBIR in the USA Effectiveness of SBIR Encouraging applied research in SMEs Providing SMEs with external management expertise Encouraging university graduates to work with SMEs A specific example of a SME–university collaboration involving graduate students An obstacle to SME–university partnering Germany Finland China encourages SMEs to engage with universities The Republic of Korea Taiwan Fostering SMEs in Singapore The example of Italy, another SME-intensive economy Chile Not-for-product development alone
9780230236516_01_prexiv.indd ix
ix
91 92 95 98 99 101 103 105 106 108 111 111 113 115 116 121 121 124 124 125 126 127 129 129 130 131 133 133
9/9/2010 2:58:35 PM
Contents
x
A synthesis from Canada Policies in favor of SMEs: room for improvement Chapter 6
Best practices for firm–university partnerships General prerequisites for firms IP policy Evaluation of universities Seed funding Public procurement Comparing Europe and North America for their performance in knowledge and technology transfer Assessing technology transfer: the case of Switzerland Technology licensing Conclusion: how can knowledge and technology transfer be optimally supported?
135 136 139 139 141 142 142 143
144 147 152 161
Chapter 7 The way forward Large firms SMEs Universities Universities: changes over the long term Caveat: we need the independent voice of universities Conclusion
163 163 170 172 180
Notes
187
Select Bibliography
193
Index
197
9780230236516_01_prexiv.indd x
183 183
9/9/2010 2:58:35 PM
LIST
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
OF
FIGURES
Ways in which firms engage with universities The incubation process for a technical venture Comparison of several OECD countries, according to the European Innovation Scoreboard (2007) Swiss rankings related to innovation and competitiveness Swiss rankings related to university–industry partnering Evolution of Swiss rankings related to university– industry partnering as a function of time Evolution of staff active in technology transfer Technology licensing and research partnering channels in Switzerland, by partner type International benchmarking of technology licensing metrics between the USA, Europe and Switzerland Average number of licenses per country for Europe Number of KTT staff per scientific publication by country in Europe Number of R&D projects with Swiss PROs reported by Swiss firms Number of R&D projects with economic partners reported by Swiss universities Public–private co-publications per million population SMEs collaborating on innovation with higher education institutions, as percentage of all firms Share of business R&D for SMEs Innovative SMEs collaborating with others, percentage of all SMEs Number of projects realized through matching grants from the government with the corresponding share of SMEs
7 83 146 148 149 150 150 151 152 153 154 156 156 157 158 159 159
160 xi
9780230236516_01_prexiv.indd xi
9/9/2010 2:58:35 PM
CHAPTER 6
Best practices for firm–university partnerships In the previous chapters, we have seen the various ways by which firms enter into collaborations with universities, in an attempt to boost their competitiveness. Let us now look, in this chapter, at what additional helpful steps are required in order to put this complex process effectively into practice within firms; we then look at some framework conditions and or policy issues, which have not been discussed yet. Then, turning to the macro level, we will compare Europe with the USA. Their performances are similar in the area of knowledge and technology transfer. China, India and Japan are forging ahead, often following their own paths. Finally, the case of Switzerland is discussed, as this country is often shown as an example for its performance in knowledge and technology transfer.
General prerequisites for firms In order for firms to engage effectively with universities, a prerequisite is for the firm to be clear about its business developments needs. The next stage is to be aware of what developments need to be carried out in-house and what should be undertaken with external partners. As indicated earlier, the tendency is for increased collaborations. Examples of criteria for this choice are: ● ● ● ●
How critical is to have full control of the development? The time horizon. Is the project more a capacity-building exercise? How specific is the development? A scenario involving planning based on geopolitical or societal trends may usually be 139
9780230236516_07_cha06.indd 139
9/13/2010 8:19:05 PM
From Science to Business
140
●
conveniently outsourced. Some issues are industry-wide, such as, for example toxicological studies of fine particles emitted by diesel engines. These must be done jointly by the manufacturers of diesel engines. What are the returns (cost and risk-sharing, but also climbing a learning curve faster) of going outside?
These criteria are specific to the firm at a given time and must be revisited occasionally. Once this map is drawn up, then the question is: what projects are most suitable for partnering with universities? Finally, what university is most fitted for a given project? Is this university available, contractually and in terms of workload? How customer-friendly is the university, its policies and its KTT office? A second prerequisite for the firm is the attitude of its staff. An open mind and high curiosity are a must, but they do not come that naturally. There is the general tendency in organizations to be inward-oriented and to consider that the internal resources of the firm are preferable, because they are rooted in the ‘culture’ of the firm. This is what is described as the ‘not invented here’ (NIH) syndrome. This can also be accompanied by the fear of discovering that outside resources may be better than anticipated. Technical experts, in particular, tend to have an ‘expert syndrome’, which make them believe in their superior competence. A well-designed and implemented management development course may act as a powerful agent of change to modify these perceptions and make the staff much more outward-oriented, proactive and curious regarding what goes on outside the firm. A cosmopolitan staff, composed of individuals from different cultures, literally brings different ‘points of view’, which are very positive for healthy debates and coming up with new ideas. Furthermore, such staff have connections with persons and institutions everywhere in the world. Thus, they promote a positive circulation of ideas and an openness to outside ones. One company appears to be very open to external ideas. It is the US healthcare and cosmetic products company Johnson & Johnson (J&J). It has a decentralized structure and an entrepreneurial perspective, where employees are on the lookout for external ideas that can lead to new J&J products. As a result, it
9780230236516_07_cha06.indd 140
9/13/2010 8:19:05 PM
Best practices for firm–university partnerships
141
is adept at ‘internalizing’ such ideas and turning them into company successes. Indeed, many of J&J’s breadwinning products, such as the supple contact lens, for example, have been brought to the company by external actors, universities, individuals or small companies. If this prerequisite of open mindedness is fulfilled, then the firm will be likely to entertain informal, but sustained contacts with the personnel from the universities most relevant to their business development needs. In a large firm, an array of ‘watchers’ regularly exchange information, making sure it is circulated and discussed, in order to best connect with the business. In contrast, in SMEs, only one or two persons watch the university scene, taking into account the ‘needs’ of the firm. In both cases, experiences are exchanged on the challenges of partnering and managing external collaborations. This is done internally, but also across firm boundaries, as much can be shared on the managerial issues without divulging what specific areas are of interest to the firm’s future business. Issues and expected trends concerning universities and public research organizations will be covered in Chapter 7. With regard to the best practices for the framework conditions, several policy issues are discussed below.
IP Policy Countries should assign the ownership of patents to the university rather than to the inventor in its staff. In this sense Germany got it right, structuring its law in this direction, whereas Italy went the wrong way by shifting the ownership to the inventor. The rationale is that, in general, universities are better equipped to extract the best value from the patent than are individuals. However, in the area of patents, it is important to provide resources for preparing and filing them. Also, it is desirable to have a coordination of the patent activity within the firm. This is why the Italian public research organization CNR, mentioned in Chapter 5, should not have decentralized the handling of patents to its individual institutes, especially in the light of the fact that no accompanying budget was provided for them to discharge this new task.
9780230236516_07_cha06.indd 141
9/13/2010 8:19:05 PM
From Science to Business
142
Evaluation of universities One fair question to ask is whether and how to evaluate universities, in particular with regards their effectiveness at transferring knowledge and technology. The UK is the country that has put in place the most systematic process in this area. Initially this was implemented by the Research Assessment Exercise (RAE), which drove the dynamic by offering increased ranking for the university if it became involved in these collaborations. This exercise did not give recognition to the knowledge and technology transfer activities. It was also resource-intensive in expert time and in management attention, even leading to bureaucracies within the universities to manage the RAE process. The latter has now been replaced by the Research Excellence Framework (REF). This is partly to make the process less costly by using metrics, rather than peer reviews. It also shifts the focus on the impact of the university’s activity. The Higher Education Funding Council for England (HEFCE) (www.hefce.ac.uk) stresses the impact and ‘effective dissemination and application’ as key criteria. The question is how undisputable the metrics used in the evaluation process will be. The REF exercise is likely to encounter some controversy, when looking at the different viewpoints of academia as to how evaluate research performance. How useful is such a tool in achieving real improvement in performance, rather than inducing undesirable dynamics? It is now planned for the REF exercise to start in 2012. Maybe, like other processes, the real value for the universities may be in going through the exercise, as a learning process, rather than in its outcome.
Seed funding Seed funding consists of providing capital to a new venture, so as to allow it to be in a position of either further funding itself or to call on external investors. This capital is usually used to pay for preliminary operations markets research or early product development. The sources of seed funding are the founders of the business, or family and friends. They may also occasionally be ‘angel’ investors. Seed capital is not a large amount of money,
9780230236516_07_cha06.indd 142
9/13/2010 8:19:06 PM
Best practices for firm–university partnerships
143
as many people start up new ventures with sums in the order of €10,000. Seed funding is different from venture capital investments, which involve higher sums, a much more complex transaction, and a corporate structure that receives the investment. Venture capital therefore comes into play when some of the uncertainties of the venture have been removed, where a reasonable business case can be made for the venture having high growth and profitability in the future. The VC industry has been badly hurt by the 2008 crisis, and this impact may last for several years. In addition, in recent times, voices have been heard to say that the VC industry has become too risk-averse, thus not fulfilling its role. Seed funding is thus a crucial link in enabling the venture to move forward, as it validates the commercial viability of the offering. It is generally considered that Europe does not have enough sources of such funding available. In contrast, the USA has a number of such sources, such as SBIR, mentioned in Chapter 5, but also a number of non-profit institutions and foundations – the Coulter Foundation, for example. The website www.foundationcenter.org lists the 50 largest foundations, which distributed close to $370 million of seedmoney grants in the USA in 2007.
Public procurement In order to create more ‘market pull’ for products and services from start-ups and SMEs, a policy encouraging the public sector to purchase these is helpful. In the European Union, roughly 16 per cent of GDP is public procurement. Such a policy provides a positive momentum. Countries such as China, the Netherlands, Singapore, and the USA have taken steps in support of public procurement to this end. Malaysia is considering such a move. In Europe, the recent EU directive on the Small Business Act, mentioned in Chapter 4, suggests similar measures. It remains to be seen how each European country will translate this into their national legislation. Such a stimulus is helpful, since large companies tend to be conservative in buying goods and services from small companies. As the advertising slogan used to proclaim: nobody has been fired for purchasing an IBM product.
9780230236516_07_cha06.indd 143
9/13/2010 8:19:06 PM
From Science to Business
144
Comparing Europe and North America for their performance in knowledge and technology transfer Europe and North America are the two most developed regions in knowledge and technology transfer. How do they compare and what are the differences between them? A ten-year lag North America developed its network of KTT offices between 1980 and 2000. This increased activity was triggered, in part, by the Bayh-Dole Act, as mentioned earlier. This gave universities more control and responsibility over the management of the intellectual property they generated. It also limited the universities to granting licenses to firms that would predominantly produce the licensed product in the USA. In Europe, due to the various approaches taken by each country, a more generalized development of technology-transfer (TT) offices took place later, between 1990 and 2000. This difference of ten years has been the main reason for the gap noted by the KTT indicators in recent times. This gap concerned the three vehicles discussed in the previous chapters: collaborative research, licensing, and university spin-outs. Recently, this gap has been steadily closing. Both regions now enjoy similar performances in terms of KTT activities. In North America, statistics about KTT have been collected by the American Association of University Technology Managers (AUTM. net) for about 25 years. AUTM obtains statistical data for about 200 universities which represent the top 10–20 per cent of the approximately 2,500 universities active in the USA. Of the top 100 universities, 96 are respondents to the annual AUTM licensing survey. In Europe, the Association of European Science and Technology Transfer Professionals (ASTP) (www.ASTP.net) has been collecting statistical information from its members since 2003. The collected data represents 10–20 per cent of European universities, but not all of the leading institutions are among the respondents. ASTP results would be higher if these universities were included. This will be the case in future surveys.
9780230236516_07_cha06.indd 144
9/13/2010 8:19:06 PM
Best practices for firm–university partnerships
145
Similar performances in Europe and the USA Despite these differences, it is possible to compare the KTT intensity between Europe and North America. Such numbers are expressed per research dollar, in order to take into account the differences in absolute research funding between both regions. The results (see below in the section on Switzerland’s technology transfer) show similar general performances. North American universities are more active in filing patents, partly as a result of higher propensity to patent, cheaper filing fees and a higher number of inventions disclosures received. On the other hand, European universities conclude more licensing deals and generate more start-ups than their American counterparts: as of 2009, 630 in Europe, as compared to 585 in the USA. As already mentioned, the number of start-ups is only one statistic. The really important figure is how many jobs these new ventures are able to create. Reliable data in this area are scarce. Overall, again, despite the fact that the statistics considered include few of the top performing universities, Europe has a strong performance in KTT-related metrics. This is less so when one tries to assess the performance of firms in commercializing the inventions licensed. As mentioned in Chapter 4, European start-ups do not grow as fast as in the USA. More generally, the lower entrepreneurial spirit in Europe results in fewer successful innovative products on the market. This is likely to explain the lower amount of royalties collected by European universities from the commercialization of their inventions and the smaller number of European companies featured in innovation rankings. One must, however, remain cautious about these conclusions, since there is no clear-cut indicator able to measure the companies’ capacity to commercialize new innovative products. One approach that should be noted is provided by the 2007 European Innovation Scoreboard, which features indicators related to innovation applications and intellectual property creation. The correlation between these two metrics results in the country rankings shown in Figure 3. Switzerland seems to enjoy strong intellectual property generation. This may be explained by the strong patent culture in the country. It is also due to the fact that this small country is
9780230236516_07_cha06.indd 145
9/13/2010 8:19:06 PM
From Science to Business
146
Innovation applications
1
0.8
Germany Israel
0.6
Finland
UK
Luxembourg Sweden Switzerland
Ireland France Belgium Iceland
Slovenia
0.4
Italy
Spain
Denmark Austria Netherlands
Norway Portugal
0.2 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Intellectual property creation
Figure 3 Comparison of several OECD countries, according to the European Innovation Scoreboard (2007)
home to the headquarters of a number of very large multinational companies, some of which are very dependant upon patents, such as Novartis and Roche. Others, such as ABB and NestlĂŠ, also file a substantial amount of patent applications. However, according to Figure 3, the capacity of Swiss firms to transform such intellectual property in innovative applications is lagging behind other European countries, such as Germany, Finland or Sweden. It is unfortunate that the USA is not included in the evaluation, but the 2007 European Innovation Scoreboard features a specific chapter comparing US and Europe performance in innovation related matters.1 Collaborative research In terms of research partnering between firms and universities, both the American National Science Foundation and AUTM report growing R&D investment by firms in university partnerships. In 2008, according to AUTM, there were $3.73 billion of
9780230236516_07_cha06.indd 146
9/13/2010 8:19:06 PM
Best practices for firm–university partnerships
147
industrial research going to universities. This represents roughly 7 per cent of the total research funding of universities, down from 10 per cent in 1999. The National Science Foundation reports $2.87 billion and 5.5 per cent in 2008. A direct comparison with ASTP survey data indicates that industry funds 10 per cent of university research activities. On a more general level, differences between US and European technology transfer relates mainly to less flexibility in negotiations in the USA. As already mentioned in Chapter 3, Europe may either assign the IP rights or license the intellectual property. This greater range of options may explain the higher intensity of license deals made by European universities, as compared with the USA. As a conclusion, despite a late start, Europe is now at the same level as North America in terms of KTT output indicators. The differences observed can be explained through cheaper patent costs and differing flexibility in the freedom to negotiation license deals. European firms seem to have more difficulty translating licensing and other technology transfer deals into successful products in the marketplace. We now turn to one of Europe’s ‘best in class’, Switzerland.
Assessing technology transfer: the case of Switzerland International rankings Switzerland is enjoying particular (but not widely recognized) success in the worldwide innovation rankings, in first place in the latest European Scoreboard Summary Innovation Index.2 In terms of competitiveness, it ranks first in the Global Competitiveness Index3 as well as fourth in the World Competitiveness Scoreboard.4 In terms of knowledge transfer and research collaboration between firms and universities, the position is also excellent with a first and second rank for the IMD and WEF respectively in 2010 (see Figure 5).
9780230236516_07_cha06.indd 147
9/13/2010 8:19:06 PM
From Science to Business
148
The progression of the Swiss position in these two rankings is impressive. In about ten years’ span, Switzerland progressed from a rank in the 10–15 range to the top spots (Figure 6). What were the main factors responsible for such a progression? In terms of practice, we see a general formalization/professionalization of knowledge and technology transfer that has occurred for the last ten years in Switzerland, mainly through the establishment of dedicated teams working for or within public research institutions (PROs). This is exemplified by the number of technology transfer professionals in Switzerland, as shown in Figure 7 (in full-time equivalent, FTE). Of course, such an increase may simply be the consequence and not the source of the intensification of knowledge and technology transfer, but clearly it has played a role in providing a clear path for companies to establish
EU summary innovation index Switzerland Sweden Finland Germany UK Denmark Austria Luxembourg Belgium Ireland France Netherlands
0.694 0.636 0.622 0.596 0.575 0.574 0.536 0.525 0.516 0.515 0.501 0.491
Source: European Innovation Scoreboard, 2009.
Global competitiveness index Switzerland USA Singapore Sweden Denmark Finland Germany Japan Canada Netherlands Hong Kong Taiwan
5.6 5.59 5.55 5.51 5.46 5.43 5.37 5.37 5.33 5.32 5.22 5.2
Source: Global Competitiveness Report, 2009–2010.
World competitiveness scoreboard Singapore Hong Kong USA Switzerland Australia Sweden Canada Taiwan Norway Malaysia Luxembourg Netherlands
100.0 99.4 99.1 96.1 92.2 90.9 90.5 90.4 90.0 87.2 86.9 85.7
Source: IMD, World Competitiveness Yearbook 2010.
Figure 4 Swiss rankings related to innovation and competitiveness
9780230236516_07_cha06.indd 148
9/13/2010 8:19:06 PM
Best practices for firm–university partnerships
149
Knowledge transfer between companies and universities Switzerland USA Israel Sweden Singapore Finland Malaysia Canada Austria Iceland Germany Denmark
7.66 7.6 7.52 6.93 6.89 6.83 6.81 6.8 6.78 6.73 6.65 6.63
Source: World Competitiveness Yearbook 2010. University - industry research collaboration USA Switzerland Finland Singapore Sweden Denmark UK Belgium Canada Germany Netherlands Taiwan
5.9 5.7 5.6 5.6 5.6 5.5 5.4 5.3 5.2 5.2 5.2 5.1
Source: Global Competitiveness Report 2009–2010.
Figure 5 Swiss rankings related to university–industry partnering
partnerships with PROs and for researchers to get clarity on their institution’s guidelines when establishing such partnerships. Deal flow and related metrics in knowledge and technology transfer Another interesting question when assessing these international rankings is to compare this subjective assessment (based on interviews with stakeholders) with a more objective series of
9780230236516_07_cha06.indd 149
9/13/2010 8:19:07 PM
From Science to Business
150
Year
Rank
1997
1999
1 2 3 4 5 6 7 8 9 10 11 12 13
2001
IMD
2003
2005
2007
2009
WEF
Figure 6 Evolution of Swiss rankings related to university–industry partnering as a function of time
Year 50
FTE(s)
40
30
20
10 2001
2003
2005
2007
Figure 7 Evolution of staff active in technology transfer Source: swiTT surveys. FTE = Full-Time Equivalent.
indicators that are based on the intensity of knowledge and technology transfer as compiled by practitioners’ professional associations such as swiTT.ch, ASTP.net or AUTM.net through their annual surveys.
9780230236516_07_cha06.indd 150
9/13/2010 8:19:07 PM
Best practices for firm–university partnerships
151
These surveys usually cover the following distinctive KTT channels: â—?
â—?
Technology licensing, which consists of the transfer through licensing/assignment of research results from the PROs to the firms. The firms (either established or just starting up) buy new technologies from universities in exchange for royalties or other financial considerations. Research partnering activities, which are established through collaborations between a firm and a PRO around a common research project. Through direct interaction, firms can get access to know how from the PROs and obtain exploitation rights to technologies that may be generated during the execution of the project.
Figure 8 shows the distribution of technology licensing and research partnering channels for Switzerland. Technology licensing activities are mostly achieved with SMEs as opposed to research partnering activities, which tend to focus rather on large firms and other public institutions. Institutions with more basic, early-stage research (such as universities) tend
Licensees of universities 2007 Public institutions 7% Large companies 31% SME 62%
Research partners of universities 2007
Public institutions 35%
SME 21%
Large companies 44%
Figure 8 Technology licensing and research partnering channels in Switzerland, by partner type Source: swiTT survey 2007, which is the most recent survey providing such a comparison.
9780230236516_07_cha06.indd 151
9/13/2010 8:19:07 PM
From Science to Business
152
to prefer to use the channel of technology licensing as compared to institutions with applied research that deploy mostly research partnering activities (such as universities of applied sciences). We will now address the specific metrics found in each of these two channels of transfer.
Technology licensing Metrics linked to technology licensing are, typically: the number of inventions made by researchers and communicated to their knowledge technology transfer office; the number of patent applications filed; the number of license deals made with companies; and the number of start-ups created which are based on such license deals (spin-offs). International benchmarking of technology licensing
%
In terms of international benchmarking of technology licensing, Switzerland demonstrates a very positive performance, as shown in Figure 9, where comparison is made between the USA (AUTM survey, 2008), Europe (ASTP survey, 2007) and Switzerland (swiTT survey, 2008). Though each survey represents only a subset of the total number of PROs in each region/country, we refer to the corresponding country/region instead of the specific survey in the text below for purposes of clarity. A more in-depth analysis 45 40 35 30 25 20 15 10 5 0 Invention disclosures
x 10
Patent applications
Executed licenses
Start-ups
Figure 9 International benchmarking of technology licensing metrics between the USA, Europe and Switzerland Note: Figures are per R&D budget ($ million): 30% ⍽ 0.3 unit for each $ million invested in R&D.
9780230236516_07_cha06.indd 152
9/13/2010 8:19:07 PM
Best practices for firm–university partnerships
153
30
20
10
0 AU BEL CH DE DEN ES FIN FR IRE IT
NE NOR PT SWE UK
Figure 10 Average number of licenses per country for Europe6 Note: n ⫽ 185.
on the representativity of these subsets and their comparability can be found elsewhere.5 Despite a lower rate of new inventions disclosed by Swiss researchers, their KTT offices are able to convert them at higher rate in executed licenses with firms and through start-up creation. Concurring results have been obtained when benchmarking Swiss technology transfer with other European countries, as shown in Figure 10. These results indicate that technology licensing is carried out more efficiently in Switzerland than in the USA and Europe. There is, however, room to increase the number of disclosures announced by researchers. Taking into account the high impact of the Swiss research,7 together with the very high intensity of basic research,8 one could reasonably expect a large pool of highquality undisclosed inventions in Swiss PROs. At present capacity, however, it would be very difficult for the Swiss TT offices to deal with such an increase. As shown by Conti and Gaulé,9 at present, Swiss TT offices would be seriously understaffed to deal with the increased flow of disclosures (assuming a direct relation between publications and disclosures) – see Figure 11. If hiring more technology-transfer professionals seems to be a solution to overcome this crunch in KTT, the question of the
9780230236516_07_cha06.indd 153
9/13/2010 8:19:08 PM
From Science to Business
154 15
10
5
0
AU BEL CH DE DEN ES FIN FR IRE IT
NE NOR PT SWE UK
Figure 11 Number of KTT staff per scientific publication by country in Europe Source: A. Conti and P. Gaulé, The CEMI Survey of University Technology Transfer Offices in Europe, working papers series, 2008. Note: n ⫽ 202.
associated costs becomes an issue. Up to now Swiss PROs have supported the vast majority of the costs needed to set up and operate TT offices (roughly CHF 50 million over the last ten years). Most of them could not draw any more resources from their research budget to further develop KTT activities and do not operate at profit in terms of KTT activities. It should be noted that we did not include in this analysis metrics linked to revenues received by Swiss PROs through the commercialization of their technologies by firms. The reason for this lies in the fact that it depends to a large extent on the ability of firms to commercialize such technologies successfully and hence does not provide a direct measure of PROs’ ability to transfer their discoveries. One can, however, note that revenues are much lower in Switzerland (about CHF10 million, according to the swiTT survey, 2007) and in Europe as compared to the USA (where it is more than $3.5 billion, AUTM survey, 2008). The high conversion rate of invention disclosures into licenses (43 per cent for the University of Geneva since inception as compared to Stanford University’s 41 per cent) seems to indicate either
9780230236516_07_cha06.indd 154
9/13/2010 8:19:08 PM
Best practices for firm–university partnerships
155
a failure of firms to commercialize such inventions or simply that more time is required to generate high revenues from them. Impact of technology licensing activities for Swtzerland Despite fewer royalties being collected in Switzerland from technology licensing activities (CHF 10 million according to the swiTT survey, 2007), using a method developed by Stanford University10 one can evaluate the impact in terms of added tax through corresponding economic growth to about CHF 50 million collected at the Swiss level (2007). The 5x multiplying factor (assuming 50 per cent of the license deals are made with Swiss companies) demonstrates that most of the added value generated by technology licensing lies outside the PROs. So why, despite even explicit OECD recommendation, has no substantial direct support been provided so far by the Swiss government to technology licensing from PROs?11 One of the main reasons lies probably in its historical development and positioning. Technology licensing activities have been funded and under direct supervision from PROs. This situation has not provided enough incentive for the government to step in due to the lack of control and leadership for its role. Another reason lies in the limited expertise at hand within the government and its pool of technology-transfer experts who are much more oriented towards research partnering than technology licensing activities. Finally, it should be noted that a similar situation occurred in the past in United Kingdom, where the government eventually stepped in to provide additional support to the PROs under the so-called ‘third-stream’ funding scheme which has now been active for the last ten years with good success.12 Research partnering The performance of research partnering of PROs with companies in Switzerland is more difficult to assess with precision. The Federal Statistical Office, together with economiesuisse, an
9780230236516_07_cha06.indd 155
9/13/2010 8:19:08 PM
From Science to Business
156 350 300 Million CHF
250 200 150 100 50 0 1996
2000
2004
2008
Figure 12 Number of R&D projects with Swiss PROs reported by Swiss firms
350 300
Million CHF
250 200 150 100 50 0 2005
2006
2007
2008
Figure 13 Number of R&D projects with economic partners reported by Swiss universities
association of Swiss firms, conduct, every four years, a survey on Swiss firms‘ R&D activities. The number of ‘extra-muros’ R&D mandates to Swiss PROs is reported in Figure 12. It is difficult to define a trend in these results and no monotonic increase in the last ten years has been reported. The situation is different in the case of the data compiled by the swiTT annual surveys (see Figure 13). The reported figures are the amount of R&D mandate
9780230236516_07_cha06.indd 156
9/13/2010 8:19:08 PM
Best practices for firm–university partnerships
157
cash allocation received by Swiss universities (including the two Swiss Federal institutes of technology) from economic partners (public and private). The comparison between both figures is difficult because of the different metrics as well as the various time spans of the surveys. Figure 12 comprises all Swiss PROs, in contrast to Figure 13, which only includes universities and the two Federal institutes of technology. On the other hand, Figure 13 not only reports the amounts collected from Swiss companies, as in Figure 12, but also those from foreign companies and public institutions. International benchmarking of research partnering International benchmarking of PROs related to research partnerships is more challenging than for technology licensing activities due to the lack of suitable indicators. As an example, PROs cannot be compared on the intensity of their partnership activities since this will depend on whether their focus is on basic or applied research and also their academic culture. Any meaningful comparison would have to be done within a group of PROs with similar profiles, such as universities of applied science, for example. On the company side, one available indicator related to research partnering is the number of co-publications of firms Switzerland
199
Sweden
128
Denmark
127
Iceland
127
Norway
110
Finland
105
Netherlands Austria
89 66
United Kingdom
61
Belgium
59
Germany
49
Figure 14 Public–private co-publications per million population Source: European Innovation Scoreboard, 2009 (based on 2007 data).
9780230236516_07_cha06.indd 157
9/13/2010 8:19:08 PM
From Science to Business
158 %
0
5
10
15
20
Finland Austria France Belgium Sweden Denmark Estonia Slovenia Netherlands Luxembourg Norway UK
Figure 15 SMEs collaborating on innovation with higher education institutions, as percentage of all firms Source: OECD STI Scoreboard, 2009.
(located in a chosen country) with PROs, per million of population. As depicted in Figure 14, Swiss firms rank first. The density of the Swiss research with regards to its population clearly provides an advantage here. It would, however, be more meaningful to measure the share of co-publications with firms in comparison to the total number of publications. This indicator is, however, interesting in the sense that it also includes not only technology but also knowledge transfer between firms and PROs. In the case of SMEs, several indicators measure the intensity of their interaction with PROs. The percentage of SMEs collaborating on innovation with higher education institutions is shown in Figure 15. Unfortunately Switzerland is missing from this ranking. It is, however, possible to estimate the percentage based on the fact that SMEs represent between 30 per cent and 40 per cent of the total amount of Swiss PROs projects with firms (swiTT surveys, 2005–7). The resulting computation gives us between 3 per cent (10 per cent x 0.3) and 4 per cent (10 per cent x 0.4). Such an evaluation positions Switzerland in second place in the tier of countries in Figure 15. This result fits with the generally poor performance of Swiss SMEs in terms of R&D intensity (Figure 16) and innovation cooperation (Figure 17), where Switzerland’s performance is below par in comparison with its European counterparts.
9780230236516_07_cha06.indd 158
9/13/2010 8:19:08 PM
159 0
20
40
60
80
New Zealand (2005) Greece (2005) Slovak Republic Norway Spain (2006) Ireland (2006) Belgium (2006) Portugal Canada (2006) Poland Australia (2006) Czech Republic Hungary Denmark (2005) Austria (2006) Luxembourg (2005) Netherlands (2005) Slovenia (2006) Korea Italy (2006) Switzerland (2004)
Figure 16 Share of business R&D for SMEs Source: OECD STI Scoreboard, 2009 (based on 2007 data).
0 Denmark Sweden Finland Belgium Cyprus Estonia Ireland Luxembourg Lithuania Iceland Czech Rep. UK Netherlands Switzerland
10
20
30
% of SMEs
Figure 17 Innovative SMEs collaborating with others, percentage of all SMEs Source: European Innovation Scoreboard, 2009 (based on 2004 data).
9780230236516_07_cha06.indd 159
9/13/2010 8:19:09 PM
From Science to Business
160 300
SMEs 250 200 150 100 50 0 2004
2005
2006
2007
2008
Figure 18 Number of projects realized through matching grants from the government with the corresponding share of SMEs Source: Annual Reports of the Commission for Technology and Innovation (CTI) Berne, Switzerland.
It is then hardly surprising that, according to an OECD economic survey, SMEs are under-represented in business–university cooperation as well as in cooperation agreements in general: they need to improve what is known as ‘absorptive capacity’,13 that is, the capacity to manage and make use of external technical know how. The lack of involvement of SMEs with PROs, as compared with other countries, is probably partly due to the low level of incentives. Switzerland is one of the few countries that does not allow direct funding of SMEs’ R&D by the government. The most widespread tool to incentivize firms (in particular SMEs) to engage with PROs consists of providing matching funds to the firm’s own R&D investment in a research partnership project with a PRO. The PRO contribution to the project (usually close to 50 per cent) is funded by the government. Figure 18 lists the number of such matching projects for Swiss firms engaging with a Swiss PRO. In 2005, the government put in place an initiative to increase collaboration between Swiss companies and PROs through the establishment of KTT consortia14 aimed at providing companies with easier access to PROs competencies (the consortia became fully operative in 2006 with CHF 12 million provided for the period 2005–7). An impact in terms of increased CTI projects is not visible in Figure 18, where the number of projects remains stable at around 250 per year. This may be due to exhaustion of
9780230236516_07_cha06.indd 160
9/13/2010 8:19:09 PM
Best practices for firm–university partnerships
161
the (roughly stable) budget made available for such measures each year. However, the number of submitted projects does not increase either (2005: 522, 2006: 407, 2007: 493, 2008: 444), which may suggest that such an initiative, rather than generating new collaborations, streamlines the process of initiation and match-making. As a consequence, other incentives may be needed to increase SMEs’ participation in collaborative projects with PROs. Another level of incentives consists of providing innovation vouchers that allow a firm to acquire R&D credits in a PRO (without the matching requisite). Switzerland has recently started to experiment with this approach. In countries such as Canada,15 the Netherlands16 or Ireland,17 the government provides R&D vouchers to SMEs to allow them to buy R&D from PROs. A review of the effect of voucher has been conducted by Cornet et al. in the Netherlands, which showed a positive effect on firms’ absorptive capacity.18 Other incentives include offering tax cuts linked to R&D investments or providing firms with direct financial support to engage in research partnerships with PROs.19
Conclusion: how can knowledge and technology transfer be optimally supported? Switzerland has achieved impressive results in technology licensing activities in the last ten years. The development and professionalization of TT offices supported by PROs, combined with the excellence of their research, has propelled Switzerland to the forefront of technology transfer world-wide. Technology licensing and research partnering form the two main channels of interaction between firms and public research institutions. Technology licensing provides good and measurable outputs in terms of deals and spin-offs generated per research budget. Technology-transfer offices are efficient in international comparison but have reached their maximum capacity with the present level of funding. Despite being almost exclusively supported by PROs, technology licensing benefits mainly the government through the collection of tax on the increased economic activity, with an estimated fivefold rate of return.
9780230236516_07_cha06.indd 161
9/13/2010 8:19:09 PM
From Science to Business
162
Research partnering’s performance is less evident. Despite positive general indicators, a deeper look points to a low intensity of participation of SMEs. Existing incentives put in place through match-making grants seem to have reached their limit, despite additional actions taken to facilitate the search for partners within PROs through setting up regional KTT consortia. Increased participation of SMEs in technology-transfer activities with PROs remains a priority. Technology licensing, with its large proportion of smaller firms, provides a good vehicle by which to strive towards this goal. It should also be further developed to fully exploit the potential provided by the Swiss research intensity and quality. In parallel, new incentives through R&D vouchers and direct R&D funding could generate a more intense participation of SMEs in partnering processes. The United Kingdom, with its ten years of experience of government support for technology transfer, could serve as a source of inspiration for future Swiss funding schemes in this area.
9780230236516_07_cha06.indd 162
9/13/2010 8:19:09 PM